It is well recognized that HIV infection significantly decreases total- and HDL-cholesterol levels in plasma. The initiation of protease inhibitor (Pl)-based highly active anti retro viral therapy (HAART) generally leads to dyslipidemia with total plasma cholesterol and triglyceride levels far exceeding the normal range. However, HDL-cholesterol level remains low in a majority of Pi-treated patients. In contrast, treatment that includes non-nucleoside reverse transcriptase inhibitors (NNRTI) nevirapine and efavirenz increases HDL-cholesterol. However, whether NNRTI-induced increase in HDL-cholesterol represents a unique property of these drugs or is simply a manifestation of a return to health remains unknown. The latter possibility would indicate that PI therapy, which does not normalize plasma HDL-cholesterol, directly reduces HDL levels in HIV infected subjects even with its reduction of viral load and health improvement. These two possibilities need to be distinguished to design the most effective treatment option to reduce the metabolic abnormalities of HAART. Another unanswered question is whether the NNRTI-induced increase in HDL is atheroprotective or whether the HDL increase is due to accumulation of dysfunctional HDL that contributes to atherosclerosis. The goals of this project are: (1) to delineate the mechanism by which HIV infection reduces HDL levels, (2) identify the pathway(s) by which restoration of health with PI versus NNRTI results in diverging effects on HDL levels, and (3) to ascertain the atheroprotective versus pro-atherogenic potentials of the HDL in patients receiving NNRTI. Specific Aim 1 will compare the fractional synthetic rates of the HDL major apolipoprotein apoA-l and the cholesterol efflux potential of macrophages and HDL/apoAl in seronegative subjects with treatment- naive HIV-infected individuals with CD4+ cell counts of >350 or <200 cells mu L to determine the impact of HIV infection and immune functions on HDL biosynthesis and metabolism. Enzymes responsible for lipidation, maturation, and remodeling of HDL will also compared to determine if differences are due to changes in expression and/or activity of these proteins. Aim 2 will compare the same HDL metabolic parameters in the HIV infected subjects after initiation of (i) atazanavir, (ii) lopinavir/ritonavir, or (iii) efavirenz therapy to identify the difference between health restoration and treatment-specific effects on HDL metabolism. Aim 3 will compare the functionality of the HDL after HAART therapy to determine if these are pro- or anti-atherogenic.